European Journal of Political Research 34: 225–247, 1998.
© 1998 Kluwer Academic Publishers. Printed in the Netherlands.
Policy convergence and policy feedback: Agricultural finance
policies in a globalizing era
1 Department of Political Science, McMaster University, Hamilton, Canada; 2 Department of
Political and International Studies, University of Warwick, Coventry, UK
Abstract. In a comparative study of five countries: Australia, Canada, the Republic of Ireland, the UK, and the USA, this article examines the degree of convergence of agricultural
credit policy content, policy instruments, and policy outcomes on a market liberal model. It
shows that all five countries have moved toward market liberal policy arrangements over the
past quarter century of globalizing and domestic fiscal pressures, but important differences
in policy remain. The Republic of Ireland and the United Kingdom move further toward a
market liberal model than do Australia, Canada, and the USA. The distinct national paths
taken to market liberalism give rise to policy feedback that hastens or retards the adoption of
a fully market liberal system. Historical choices of policy instruments and path dependence
help account for continuing policy divergence.
By examining policy convergence in agricultural finance policies in five countries and drawing on the concept of policy feedback, this article sets out a
nuanced version of the argument why national differences persist. Following
Bennett (1991), policy convergence embraces changes in policy objectives,
policy content and instruments, and ultimately policy outcomes. Based on
this understanding, policy convergence occurs in agricultural finance policies,
but it is uneven, and cannot be explained in terms of differential time lags.
Agricultural finance policies would seem a likely area of policy convergence because of the growing commitment of policy makers to market liberalism. Members of the OECD began working on a set of principles in the early
1980s that culminated in the adoption of a common objective in 1987: ‘to
allow market signals to influence by way of a progressive and concerted reduction of agricultural support, as well as by all other appropriate means, the
orientation of agricultural production’ (OECD 1991: 213). Adoption of this
policy objective could conceivably lead over time to common policies with a
similar market liberal content, delivered through similar policy instruments,
to wit private financial services firms. Over time, such convergence on content
and instruments should lead to common outcomes, where private firms come
to dominate the provision of agricultural finance. This dominance of private
firms, in turn, likely contributes to a widening of the gap between large and
small producers and possibly to encouragement of intensive agriculture over
more ecological approaches.
In his review of Paulette Kurzer’s work, Zysman (1994: 803) argues that
convergence will evolve following a distinctly national path. Feick’s (1992)
review of institutional constraints would imply such a process as well. Certainly, the series of comparative studies reviewed by Unger & van Waarden
(1995) plus van Waarden’s own work (1995) that look carefully at the constraints placed on change by the give national institutional repertoire would
support Zysman’s argument.
Without denying the importance of the kinds of macropolitical variables
identified by these authors in exploring variations in policy convergence, we
would like to stress the importance as well of micro-level analysis of policy
content and policy instruments. It has long been a tenet of policy studies
that policies create politics (Lowi 1972). In his recent work on policy retrenchment, Pierson (1993, 1994) has expanded on this idea with his concept
of policy feedback. Policy feedback refers to the resources and incentives
created by policies themselves to group formation and activity, and to the
processes of social learning that follow from the implementation of policies.
Among other factors, Pierson stresses that policies may offer incentives to
individuals and groups that ‘lock in’ certain paths of policy development.
‘Policies may create incentives that encourage the emergence of elaborate
social and economic networks, greatly increasing the cost of adopting oncepossible alternatives and inhibiting exit from a current policy path. Individuals make important commitments in response to certain types of government
action. These commitments, in turn, may vastly increase the disruption caused
by new policies, effectively locking in previous decisions’ (1994: 44).
Pierson adds that policies that create high levels of interdependence between individuals and organizations, particularly when this interpendence
stretches over long periods, are most likely to create this path dependence.
It follows that variations in the degree to which policies create ‘lock in’ effects may be associated with variations in the degree of policy convergence.
Scharpf (1991: 10–12) has stressed that the ‘accidents’ of the policy repertoire
already in place at a time of change will lead to rather different responses by
nation-states. We suggest four aspects of policy instruments and policy implementation that help account for variations in policy convergence: whether the
instrument creates direct contacts between producers and the state, the degree
of structural similarity between state financing agencies and private firms,
the strata of farmers served by state agencies, and the length of time state
agencies have been active. In this respect, policy feedback becomes another
element that accelerates or stymies policy convergence.
Our goal here is a modest one. We highlight aspects of policy instruments
and policy content that lead to more or less ‘lock in’. We show that these
variations offer the possibility of credible explanations for variations in policy
convergence. The paper develops this argument in three steps. The first sets
out the key characteristics of policy systems for agricultural credit in five
countries: Australia, Canada, the Republic of Ireland, the UK, and the USA.
The second section traces the changes in agricultural credit policies since
1970 in the five countries. The third section of the paper examines some
possible explanations for the differences in agricultural credit policies that
Types of agricultural finance regimes
In discussions of farm finance, analysts normally distinguish among three
distinct credit markets. Throughout the first half of this century, if not longer,
these credit markets did not fit well into the usual operations of commercial
banks in their relations with industry. Accordingly, often in cooperation with
agricultural producers, states acted to create specialised institutions. Farmers
needed long-term mortgage credit to purchase and develop land, a need not
readily met by the negotiable long-term bonds favoured by industrial corporations and underwritten by banks (Schickele 1954: 78). They also required
intermediate credit for terms of three to five years, the time required to liquidate investments in livestock, equipment and facilities. Private commercial
banks, however, often looked askance at intermediate credit, preferring to
match assets to their short-term liabilities (deposits) and thus to offer shortterm credit of 60 to 90 days to industrial and trading firms. And when agricultural producers needed short-term operating credit, banks’ conventional risk
analysis developed for assessing their corporate clients led them to conclude
farming was a high risk business. Thus, they avoided farmers altogether or
offered them loans at high interest rates, with little flexibility for renewal or
extension in response to changing agricultural growing conditions.
Farmers and governments responded to these financial obstacles by forming private co-operative agricultural banks, or co-operative banks backed by
a state guarantee, or state-owned lenders. Most of these institutions came
into existence prior to 1940. After 1940, they assumed greater importance as
credit needs grew and governments took a greater interest in the structure of
In the postwar period, the state intervened extensively to promote the
growth of productivist agriculture. Various market regulations and price and
income supports protected and stabilized farm incomes. Building on this
stability, states then used selected financial incentives to encourage suitable
farm structures – more land, new machinery, more use of chemicals, modern
facilities. Such structural intervention came most often through the manipulation of credit (OECD 1970, Vol. 1: 81; Green 1987: 666). With credit in
demand and with various government credit agencies already in place because
of capital shortages earlier in the century, governments were able to intervene
readily to direct agricultural development.
Since 1970, a number of international and domestic changes have created
pressures on states to reduce their protection of agricultural incomes and
to withdraw from structural intervention. In the income support arena, the
combination of the insertion of agriculture into the world trading system during the Uruguay Round negotiations and acute domestic fiscal problems put
intense pressure on costly farm programmes. Income supports were largely
ended in Australia in the 1980s (Coleman & Skogstad 1995), reformed substantially in the EU with the changes in the CAP in 1992 (Fouilleux 1996;
Grant 1995; Rieger 1996), and significantly decoupled from production in
the 1985, 1991 and 1996 Farm Bills in the USA and related reforms in Canada
(Coleman, Atkinson & Montpetit 1997; Coleman, Skogstad & Atkinson
1996). In the structural arena, postwar national systems for protecting financial services were liberalized, resulting in the desegmentation of many
credit markets (Cerny 1993; Coleman 1996). Long a market segment with
significant state oversight, agricultural finance arrangements became subject
to similar pressures.
When the arrangements that were put in place during the immediate postwar period are contrasted to some of the market liberal proposals of the 1990s,
three types or models of agricultural finance can be distinguished. The core
dimensions of these models are summarized in Table 1.
In the developmental model, states intervene to occupy market niches of
little interest to commercial banks (usually long-term or intermediate credit).
They also provide lender of last resort facilities for poorer producers who are
unable to find credit from other lenders. Some, but not all, states also define
a series of structural objectives for agriculture. All states will tend to operate
through state-supported agencies to provide direct payments or allocate credit
under more favourable conditions to producers who act consistently with
those objectives. In these various activities, states often subsidize the cost of
loans or help to lower interest rates by guaranteeing loans made by state agencies or private firms. Normally, the state will own one or more corporations
whose mandate will be restricted to financing agricultural producers. These
agencies usually operate outside domestic banking laws and regulation.
Table 1. Classification of agricultural finance systems
Mixed market
Market liberal
State role
selected market
competition in
selected market
market framework
Social support
support for
Providing support
on an ad hoc
basis for
No specific
support for
Principal policy
lending agencies
agencies and
private firms
Private firms
interest rates
Loans backed by a
state guarantee
interest rates
Mandate of statesupported agencies
producers only
Broader agri-food
No restrictions
on business
activities or no
The market liberal model operates on assumptions opposed to the developmental one. Policies are devised based on the belief that agriculture is an
economic sector like all others and that its credit needs can be met by the
conventional banking system. In this model, therefore, lending takes place at
market-determined interest rates, with loans being sold by institutions defined
as banks by domestic banking laws. If state-owned corporations remain active
under this model, they have the legal status of banks and operate as commercial entities. Like other banks, their mandate will not be restricted to financing
agricultural producers only, but will extend to all commercial opportunities.
Finally, we note an intermediate or mixed market model that retains a
few features of developmental models, while assuming more of a market
liberal focus.1 In this model, state-supported agencies continue to serve particular market niches, but in competition with commercial banks. In part,
this business is a natural consequence of building up a clientele during a
developmental model phase. But their continued presence often rests on producers’ beliefs that banks are not firmly committed to serving farmers and
that state-supported lenders will keep these banks ‘honest’ and ‘competitive’.
Such state agencies also provide reassurance that an instrument is available
to provide ad hoc support to producers experiencing severe economic stress.
Also characteristic of a mixed market model is a widening of the business
mandate of state-supported agencies beyond agricultural production per se to
the larger agri-food economy.
Convergence on market liberalism?
Policy content and policy instruments
Based on these ideal types of agricultural finance regimes, an ordinal index
can be devised for assessing proximity to a given model. This index is based
on the five dimensions summarized in Table 1. When arrangements in a country came closest to the developmental model, the country was assigned a 3.
Arrangements approximating best the mixed model were given a 2 and those
most consistent with the market liberal model were scored 1. Thus a close
match with the developmental model obtains a score of 15, with the mixed
model 10, and with the market liberal model 5. The results of this indexing
procedure are provided in Table 2.
In 1970, the agricultural credit regimes in Australia and Canada approximated best the developmental model. Those in Ireland and the USA had
elements of both the developmental and mixed models. The UK already featured arrangements that matched the mixed model. Over the course of the
ensuing quarter century, Australia moved closer to the mixed model, placing
it at a ranking the same as the USA where arrangements changed the least.
Canada’s policy content and instruments changed more, bringing them to the
liberal side of the mixed model. Finally, both the Republic of Ireland and the
UK experienced sufficient changes to bring these countries to a clear market
liberal regime by 1995.
Australia. Three types of private firms have been agricultural lenders in Australia: the major commercial or ‘trading’ banks which dominate banking markets, smaller finance houses termed pastoral finance companies, and life insurance companies. In 1970, state involvement in agricultural finance featured first the ownership of the Commonwealth Development Bank (CDB)
which had a mandate to provide long-term finance in situations where ‘finance would not otherwise be available on reasonable and suitable terms and
Table 2. Country rank orderings, agricultural finance finance models
State role
Mandate of
Total score
1970 1995
1970 1995
1970 1995
1970 1995
1970 1995
conditions’ (Gifford et al. 1967: 165; Australia 1981: 79). This lender of last
resort facility reflected a commitment to social support for marginal producers. With gaps remaining in the long-term market, however, the government
agreed to proposals from private bankers to create a wholesale refinancing facility, the Primary Industry Bank of Australia in 1978 (Ives 1987: 31). Jointly
owned by the Commonwealth government, a consortium of state banks, and
the major trading banks, the PIBA used funds provided by the Commonwealth at a concessional rate, plus capital raised on financial markets, to
channel long-term loan capital to private and state-owned lenders (Australia
1981: 708).
Noting a further gap in intermediate lending, the trading banks had been
encouraged to enter these markets with the creation of the Term Loan Funds
in 1962 and Farm Development Loan Funds in 1966 (Standen 1982). These
funds enabled the trading banks to offer intermediate loans at concessional
rates, accounting for close to one-quarter of farm indebtedness in 1980. Finally, in 1971, consistent with a developmental model, the Commonwealth
government created the Rural Reconstruction Scheme (Stent 1981; Watson
1989: 163–164).2 Jointly administered with such sub-national state-owned
lenders as the Rural Finance Corporation of Victoria or with sub-national
distribution agencies, the Scheme either provided guarantees for loans by
private lenders or enabled state financing authorities to make direct loans at
concessional rates. These loans were to assist farmers to overcome financial
difficulties, to improve their productivity, or to exit the sector (Jenkins et al.
1983; Musgrave 1990).
Beginning in the late 1980s, Australia’s arrangements began to shift away
from the developmental model. First, in 1988, the Primary Industry Bank
was taken over by the Rural and Industries Bank of Western Australia (now
WestBank), a trading bank owned by the government of Western Australia
(Spillman 1989: 251). WestBank has a broad business mandate and has used
PIBA to expand its business in the agricultural and resource sectors. Second,
as we shall note below, the importance of the lender of last resort, the Commonwealth Development Bank, has declined consistently since 1980. Third,
the intermediate lending market has matured, leading to a decline in subsidised loans. Finally, in 1992, changes to the RAS signalled a shift away from
supporting the marginal producer to targeting viable farmers needing productivity and efficiency improvements only (Maslen 1993). If such support were
to be completely withdrawn, Australia would fit fully the characteristics of
the mixed model.
Canada. Canadian policies had produced a system that lay quite close to the
developmental model. In 1944, concerned about the ‘modernization’ of agricultural production, the federal government chose the guaranteed loan policy
instrument and passed the Farm Improvement Loans Act to encourage banks
and credit unions to enter the intermediate domain. In 1980, FILA/FIMCLA3
loans accounted for about 12 percent of intermediate credit outstanding.
Provincial government agencies also became active in the intermediate markets. In 1959, the federal government intervened further when it passed the
Farm Credit Act, creating a self-governing, state-owned corporation, the Farm
Credit Corporation (FCC). Drawing on allocations from the Consolidated
Revenue Fund of the federal government, the FCC was authorized to make
long-term loans, a field largely abandoned by private institutions. Loan rates
tended to be set close to what was offered by private individuals, and thus
probably slightly below market rates (although it is difficult to assess market
rates when private financial services firms were not present). Provincial agencies such as the Office du Crédit Agricole in Quebec, created in the 1930s,
also offered long-term loans at subsidised interest rates.
Although a social role had been implicit to the mandates of government
agencies earlier, this role became even more explicit following difficulties in
the early 1980s experienced by cereal grains sectors of Canadian agriculture
based in the Prairie Provinces. In response, the FCC was asked to assist marginal producers by taking on a lender of last resort role with the introduction
of the Special Farm Assistance Program in 1981. This social use of the FCC
was costly, requiring the federal government to bail out the corporation by
converting 400 million dollars of its debt to contributed capital in 1987.
These costs were an important catalyst to policy changes that signalled
moves toward the mixed market model. In 1988, the federal government
excused the FCC from its lender of last resort role and asked it to provide
mortgage credit on a break-even basis. Consequently, interest rate differences
between the FCC and private lenders declined. Additional amendments in
1993 to the FCC Act expanded the corporation’s mandate to move it toward the mixed market model; it was now permitted to provide financing
for on-farm diversification and value-added operations beyond the farm gate.
Similarly, after a reorganisation in 1994, the OCA, renamed the Société de
financement agricole, was given a new mandate that permitted it to offer
loans in support of value-added activities in the broader agri-food sector. In
short, by the mid-1990s, government-owned corporations had moved from
a position of close to exclusive domination of long-term markets to one of
direct competition with private firms.
Republic of Ireland. As Table 2 indicates, Irish arrangements for agricultural credit moved from a developmental model to a market liberal model.
Government intervened first in agricultural credit with the formation of the
Land Commission in 1870 to facilitate the purchase of land by tenant farmers
(Grant & MacNamara 1994: 19). After the land transfers had been largely
completed and Ireland became independent, Irish farming entered a severe
financial crisis in the early 1920s. Government investigation revealed that the
credit sources available to small farmers were inadequate prompting the creation of the Agricultural Credit Corporation (ACC) in 1928 to fill a long-term
lending niche. This state-owned facility, however, did not become very active,
lending out a little over 10 million pounds over the next 33 years (Dooney
1988: 175). The low use of credit Sources such as the ACC by Irish farmers
was due both to bureaucratic obstacles and the low level of farm income prior
to 1970 (Atwood 1983: 13).
Beginning in the 1970s, however, in anticipation of entry into the European Economic Community, the government introduced loan schemes for
intermediate credit to encourage adaptation to Community conditions. The
ACC and the commercial banks were both used as the instruments for the
implementation of these Programs. Throughout the 1970s, the commercial
banks became progressively more active in both the long-term and intermediate markets, entering thereby into increased competition with the ACC.
This increased activity is hardly surprising. The Common Agricultural Policy
provided a measure of income security not present before. Ireland became
eligible for socio-structural support to modernize its farms. This support is
delivered through the banking system, often with credit being subsidized.
Reflecting this trend toward market liberalism, the government changed the
ACC’s mandate in 1988, permitting it to have 25 per cent of its business in
areas outside farming (Bolz et al. 1993: 162). In 1992, the market liberal
transformation was completed with the conversion of the ACC into a regular
bank under Irish banking law, with a broad business mandate.
United Kingdom. State intervention in agricultural finance has been relatively
restrained in the UK. British commercial banks had well-established business
relations with agricultural producers, particularly those owning larger farms.
During the 1920s, however, concerns were raised about the availability of
long-term mortgage credit for smaller tenant farmers wishing to own their
own land. The already strong position of banks in the agricultural credit
policy community was evident when a solution to this problem was found,
the establishment of the Agricultural Mortgage Corporation (AMC) and its
Scottish counterpart, the Scottish Agricultural Securities Corporation (SASC)
in the late 1920s (Grant & MacNamara 1995).
In effect, these agencies were instruments designed to provide further security for commercial banks in the long-term lending field. The AMC’s share
capital was distributed among the Bank of England and the leading clearing
banks (Bolz et al. 1993: 101). Similarly, SASC’s shares were held by the
Bank of Scotland and other Scottish banks. The AMC received a grant and
an interest free loan from the Treasury. With its parliamentary charter, it was
able to raise additional funds on the London Stock Exchange (OECD 1970,
Vol. 2: 40). The AMC and SASC did not compete with the banks. They were
controlled by them, raised funds for long-term and intermediate-term lending,
and were provided an indirect government guarantee for these loans (Grant
& MacNamara 1994: 16). They were not registered banks under UK law, but
operated under the supervision of the Ministry of Agriculture and Food.
This public-private arrangement consistent with a mixed market model
came to an end in the early 1990s when the AMC was purchased by Lloyds
Bank, a major clearing bank. Although the SASC continues to operate, it is
no longer taking new loans. Its market share is so small (1.5 per cent of total
bank long and intermediate lending in 1990) (Bolz et al. 1993: 102), we can
only conclude that the UK fits very closely the market liberal model, with
commercial banks in a dominant position.
United States of America. The arrangements for agricultural finance that were
put in place in the USA during the first half of the twentieth century fit best the
developmental model. Governments intervened to fill market gaps in the longterm and intermediate credit markets, primarily through the government-
sponsored, but private co-operative based, Farm Credit System (FCS)
(Peoples et al. 1992: 13). With the creation of the Farm Credit Administration as an oversight body in the 1930s, the FCS took the form it was still
to have in 1970: Land Banks, the Intermediate Credit Banks, the Banks for
Cooperatives and the Production Credit Associations. Beginning in the 1930s,
the US government also committed itself to addressing the credit difficulties
of smaller, poorer farmers, a commitment acted upon through the Farmers
Home Administration (FmHA). In 1970, the FmHA was continuing to serve
marginal producers in the farm community in all three market niches as a
lender of last resort (US Department of Agriculture 1985: 14).4
As Table 2 suggests, the US has not shifted very much toward the mixed
model. The most evident change is one of increased reliance on private firms
for providing credit, a change indicated by two developments. First, the 1985
Farm Bill decreased FmHA appropriations for direct lending, while increasing those for guaranteed lending. The Omnibus Budget Reconciliation Act of
1990 continued this trend (US General Accounting Office 1992: 16). Guaranteed lending rose from 35.9 per cent of farmer program obligations in 1986 to
68 per cent in 1993. In essence, the FmHA gradually shifted its role from one
of occupying markets avoided by private firms to one of providing backing
for loans made by commercial or FCS banks.
Second, changes were made to the privately-owned, but governmentsponsored, Farm Credit System in the wake of financial crises in the 1980s.
Congress restructured the FCS to provide it with a more market-oriented internal structure and with regulatory oversight more similar to other financial
institutions. Like the FCC in Canada and the ACC in Ireland, the FCS also
sought to expand its mandate to include the broader agri-food sector or all
business lending in rural areas (Freshwater & Riemenschneider 1994). Its
argument that it is vulnerable in remaining a ‘single sector lender’ signals a
desire for further movement toward a mixed model, a desire not yet agreed to
by Congress.
Policy outcomes
As Bennett (1991) has argued, policy convergence extends beyond movement
toward common policy content and policy instruments to similar policy outcomes. As policy arrangements move away from the developmental model
toward the mixed and market liberal models, we would expect state agencies
to cede the marketplace to private financial services firms. In addition, if we
observe variation in the degree of movement toward market liberalism, then
we should also observe variation in the relative change in market shares held
by private firms. The further arrangements move toward the market liberal
model, the larger should be the market share of private firms.
Table 3 presents some data that relate to this hypothesis. The two states
where convergence on the market liberal model was highest, Ireland and
the UK, also show a trend to private firm dominance of agricultural credit
markets. By 1993, private firms hold 84.8 percent of the Irish market and
98.5 percent of the UK market. In Table 2, Canada also appeared to move
fairly definitively away from the developmental model. Table 3 shows that
this movement has been accompanied by a secular increase in the market
share of private firms.
The two countries which showed the least movement away from the developmental model, Australia and the USA, also did not have a definite shift in
policy outcomes. In both countries, private firms held a larger market share in
1970 than they did in 1993. Both also have a pattern where private firms lose
market share in the debt crisis years of the early to mid 1980s, but regain share
as the crisis eases. Consistent with the developmental model, the activities of
state agencies increase in Australia and the USA when economic difficulties
increase for some subset of farmers.
In summary, there is some evidence that policy outcomes in agricultural
finance track convergence in policy content and policy instruments. The more
market liberal changes are in the latter, the stronger the position private financial services firms have in agricultural credit markets. Such a shift has
a number of other possible important implications for agricultural structures
and policies. There are distributional implications in terms of farm size. If
larger farmers have better access to credit because they are seen as more
market-oriented by lenders, they are better able to purchase land to expand
their operations than smaller competitors. In addition, a market liberal system
of agricultural finance is likely to encourage more intensive forms of farming.
If as our research suggests, lenders have an image of what they see as viable
(commercial, profit-oriented farmers), they are likely to prefer those forms of
farming which offered an assured return in the short to medium run. Thus,
without any ideological preferences about the merits of intensive versus organic farming, they may be reluctant to fund a transition to organic farming.
The expectations of a return on the investment may be too long-term for the
private lenders’ liking.5
National differences and policy feedback
The analysis of policy change traced above shows a general trend toward
the market liberal model in all five states. It also reveals that this policy
convergence is by no means absolute. The five states started the present era
with different policy objectives, some more liberal than others. In addition,
only two states, the UK and Ireland, have evolved to match fully the market
Table 3. Farm indebtedness to financial institutions, 1970–1993
1970 1975 1980 1985 1990 1993
Commercial banks: Term and farm development 10.1 16.7 24.1 24.3 9.8 16.7
Private: Commercial banks, pastoral finance, life 60.7 48.8 38.0 33.7 51.2 51.1
insurance companies
Commonwealth Development Bank
8.4 9.5 7.8 8.0 6.4 4.0
Other govt. agencies including state banks
20.6 24.8 24.7 23.7 27.5 23.5
Primary Industry Bank of Australia
5.7 10.1 4.9 4.5
Commercial banks: FILA/FIMCLA
10.1 7.6 4.1 2.3 1.5
Private: Commercial banks, Credit
37.8 44.5 59.5 59.2 61.8 63.9a
Unions/Caisses populaires; life insurance, trust
and loan companies
Federal governmental agencies
38.8 34.3 25.4 24.7 18.6 17.7
Provincial governmental agencies
13.1 12.5 10.0 13.9 17.9 17.2
Agricultural Credit Corporation
Private: Commercial (licensed and associated)
United Kingdom
Agricultural Mortgage Corporation/Scottish Agricultural Securities Corporation
Private: Commercial banks
United States
Farm Credit System
Farmers Home Administration
Private: Commercial banks,
44.8 37
55.2 63
29.3 23.3 15.2b
70.7 76.7 84.8
57.8 36.2 28.7 32.6
42.2 63.8 71.3 67.4 98.5
33.7 40.4 44.1 41.2 32.4 32.3
8.9 7.3 14.5 18.0 15.5 11.1
life insurance 57.4 52.2 41.1 40.9 52.1 56.6
a Includes FIMCLA fluids.
b 1992 figures.
Sources: Bureau of Agricultural and Resource Economics, Quarterly Review of the Rural
Economy, Agriculture and Resources Quarterly; Statistics Canada, Agricultural Economic
Statistics, Catalogue No. 21-603B, May 1994; Farm Credit Corporation, Farm Credit
Statistics, various years; US Department of Agriculture, Economic Research Service,
Situation and Outlook: Agricultural Income and Finance, various years; Agriculture in
the UK, Farm Incomes in the UK; Agricultural Credit Corporation and Central Bank of
Ireland, Quarterly Bulletin; Matthias Bolz et al., Agrarkreditsysteme in der Europäischen
Union (1993).
liberal model. The other three states have policy objectives and instruments
that correspond better to the mixed market model.
As we indicated in the introduction, policy feedback alerts us to the possibility that particular characteristics of policies themselves may channel policy change. Policy convergence, therefore, may be shaped by national institutional repertoires and by the properties of policy instruments and the
approaches to policy implementation. In examining policy change in agricultural finance, four characteristics of policies contribute to variations in policy
1. The nature of contracts created between farmers and the state. When
a state-owned credit facility lends directly to producers, it may become
viewed by them as a resource. Farmers get accustomed to doing business with the agency and organize politically to preserve it. In contrast,
where contacts are at best indirect such as through refinancing facilities
or guaranteed loans, less policy lock-in will occur.
2. The length of time state agencies have been active. The longer state agencies have provided benefits to farmers, the more valued will be the agency
and the more habitual will be farmers’ links to it.
3. The degree of structural similarity between state agencies and private
firms. The more similar the structures and activities of state credit institutions are to those of private firms, the more they are likely to share in the
‘banking culture’ and the easier privatisation will be. In contrast, significant structural differences will contribute to lock-in to the developmental
4. The strata of agricultural producers served by the agency. If state agencies serve primarily marginal, less credit-worthy farmers, these agencies
become unattractive take-over targets for private firms. If a state agency
serves competitive producers, its loan portfolio becomes more attractive
to private lenders and less of an obstacle to the market liberal model.
The importance of differential policy feedback based on such policy differences is shown in the following three-step analysis. First, we look at a
comparison of Ireland and the United States. Both of these states had arrangements that positioned them between the developmental and mixed market
models in 1970. A quarter century later, Ireland had evolved toward the market liberal model while the USA moved slightly closer to the mixed market
model. Second, we examine Australia and Canada, both of which started even
closer to the developmental model, but evolved in varying degrees toward the
mixed market model only. Finally, we comment upon the United Kingdom
which was already at the mixed model in 1970 and then moved to a market
liberal position. Our focus is primarily on the policy instruments already in
place in 1970, and on the institutional obstacles they pose in their own right
to market liberal change.
Ireland and the United States of America. Given that both countries had
arrangements that placed them at a ranking between the developmental and
mixed models in 1970, why did Ireland move further toward market liberalism than the USA in the following years? Politically, Irish governments were
not as neo-conservative during the 1980s as was found in the USA. Studies
of agricultural interest associations do not speak about them having a highly
market liberal orientation during this period (Collins 1995). In fact, we would
be surprised if any analysis showed these associations as conservative as the
largest American general farm organization, the American Farm Bureau Federation. More promising avenues of explanation come from an examination
of the policy communities and policy instruments in the two states.
First, a policy community focused on farm finance arose, composed principally of state officials and state agencies, and became institutionalized earlier in the USA than in Ireland. American agriculture moved toward the productivist model throughout the first half of the twentieth century. Credit issues
forced state intervention as early as 1916 and the state agencies that evolved
over the next two decades, the Farm Credit System (FCS) and the Farmers
Home Administration (FmHA), came to occupy central places in key credit
markets and in the sub-government of the agricultural finance policy community (Hamilton 1991). Commercial banks, in contrast, were only to move to a
key position in the policy community after the Second World War.
Ireland adopted the productivist model much later than the USA. Economic change only began in earnest in the 1960s, and then really accelerated
once Ireland was poised to enter the European Economic Community. Although the Agricultural Credit Corporation had been created in the 1920s,
demand was never high and it remained very small, thereby giving it shallower roots in the farm community than its US counterparts (Atwood 1983).
Its role became more important during the 1960s, but was shared by this time
with the commercial banks, themselves dominated by the Bank of Ireland –
AIB duopoly. With agricultural credit being a joint venture of the agricultural
and banking policy communities, the ACC never had as powerful a position
in the policy process as its American counterparts.
These historical differences meant that several generations of US producers developed working relationships with the FCS and FmHA, while Irish
farmers had much less experience with the ACC. American producers thus
tended to see their agencies as potential bulwarks against raw market forces.
Although not always happy with their lending policies, farmers did see them
as more of ‘their own’, and as a means for keeping the private banks ‘honest’.
This interdependence and political support created a more ‘locked in’ policy
path than in Ireland.
This difference in historical experience and thus political support for state
agencies must be coupled to important differences in structures of these credit
facilities the policy instruments themselves. In 1970, the ACC had three
sources of funds: government shares, advances from the state treasury, and
deposit accounts (OECD 1970, Vol. 2: 40). In accepting and managing deposits, the ACC was structurally similar to commercial banks and financial
co-operatives. It had to meet the same regulatory requirements as commercial
banks which further lowered the internal ‘cultural’ differences between the
ACC and privately-owned commercial banks. In short, the characteristics of
the policy instrument chosen in Ireland led to fewer structural and ‘cultural’
barriers with commercial banks than exist in the USA.
In contrast, funding sources were less diversified for the US agencies. As
a ‘government-sponsored’ private co-operative, the FCS relied primarily on
the sale of government-backed bonds to raise money for lending. The FmHA
drew its capital directly from advances from the state. Neither institution took
in deposits, leading to important differences in regulatory arrangements and
internal culture from private firms. The FCS was self-regulated, with the Farm
Credit Administration being the oversight body, a system roundly criticized
for being too cosy during the farm debt crisis of the 1980s (Freshwater 1989).
FmHA was supervised by the US Department of Agriculture, supplemented
by rather frequent attention from the General Accounting Office of the US
Congress. Neither institution received any attention from the national banking
supervisors: the Federal Reserve, the Federal Deposit Insurance Corporation,
or the Comptroller of the Currency.
With its greater similarity in structure and regulatory culture to private
firms, it was easier for the Irish government to convert the ACC into a regular
bank, a move welcomed by ACC management, and to place it directly under
the supervision of the Bank of Ireland. But there was no easy path to follow
to convert either the FCS or the FmHA into regular banks. Lacking a key
aspect of banking business, customer deposits, and organized as a network of
co-operative banks, the FCS was not an obvious or attractive takeover target.
Moreover, with a separate regulator and a culture distinct from commercial
banking, FCS management was highly predisposed to resist such change. The
FmHA acted essentially as a lender of last resort, compiling in the process a
loan portfolio of the leftovers from the FCS and the commercial banks. Why
would any self-interested bank want to purchase such a loan portfolio? With
such high structural and cultural obstacles and with the deeper support they
enjoyed in the agricultural policy community, American politicians, even if
enamoured with market liberal ideology, were bound to think twice about
spending political capital on converting the FCS or FmHA into commercial
Australia and Canada. Both Australia and Canada entered the 1970s with
policy arrangements that approximated best the developmental model. Both
have evolved in 1995 toward the mixed market model, with Canada proceeding further. The principal difference between the countries is the continuation
of a social support role in Australia. Before turning to this difference, we look
first to common properties of policy instruments that helped lock-in to a more
developmental approach.
First, the state institutions in both countries had been in place and active
for a long period of time. The predecessor to the FCC in Canada, the Canadian Farm Loan Development Board, was set up in 1927 (Canada 1979);
at the provincial level, the Office du Crédit Agricole, was created in 1936.
In Australia, the predecessor to the Commonwealth Development Bank was
established in 1943 (Ives 1987: 15). At the state level, the Rural and Industries
Bank of Western Australia had been founded in 1889. Accordingly, producers
had time to develop the ‘habit’ to bank with such government agencies and
have argued they were needed as a counter to private banks.
Second, these agencies generally had a different structure from private
banks. The CDB and various state lenders in Australia and the FCC and
provincial lenders in Canada did not accept deposits and were funded either
through loans from the exchequer or through government-backed bonds. As
such, they were not easily convertible to commercial banks. Where stateowned corporations had structures more similar to commercial banks, the
Primary Industry Bank (a refinancing corporation partially owned by the trading banks) and the Rural and Industry Bank of Western Australia (accepted
deposits) being notable examples, governments were able to integrate these
into the regular banking system.
The structure of the Australian and Canadian federal systems reinforced
the ‘lock in’ that came from historical experiences and structural dissimilarities. Federalism has encouraged the development of more autonomous
and complex agricultural policy communities at the sub-national level than
in Ireland and the UK and even than in the USA (Coleman & Skogstad
1995; Skogstad 1990). In the Australian state of Victoria and the Canadian
province of Quebec for example, sub-national governments had set up their
own agricultural lenders. These lenders received strong support from farm organizations in the respective jurisdictions, thus making them more politically
difficult to ‘reform’ or to ‘privatise’. Farm organizations at the sub-national
level found it somewhat easier to mobilize their members in favour of such
institutions because production tended to be more homogeneous in one re-
gion than it was nationally. In such multi-tiered systems, the opportunities to
block market liberal reforms were more plentiful. Accordingly, sub-national
governments like those of Victoria and Quebec were less pressured to remove
some agencies and programmes that were more developmental in character. In Quebec, nationalism also played a role as the OCA was seen as a
francophone Quebec organization in competition with the more Anglo FCC.
Neither these national institutional variables nor policy feedback explain
well, however, why Australia retains a stronger social support component
than Canada (see Table 2). The explanation probably lies in other areas of
agricultural policy. During the 1980s, Australia lowered significantly its financial support to agriculture, rendering its farmers the least subsidized in
the OECD save New Zealand. Canada, in contrast, has retained higher levels
of protection. The rapid withdrawal of support in Australia has necessitated
some government involvement in structural adjustment and the continuation
of lender of last resort facilities for farmers in most difficulty.
The United Kingdom. The movement of the United Kingdom from the mixed
model to the market liberal model was not a surprising outcome.The early
development of commercial agriculture in Britain was one of the factors that,
through the availability of surplus capital, facilitated its well-known head start
in industrialisation. Increased international competition in the last quarter of
the nineteenth century led to the disappearance of many marginal holdings.
Thus, by the first quarter of the twentieth century, small scale subsistence
farming had largely disappeared except from some remote fringe areas (with
the Scottish crofters being protected by special legislation).
Hence, agricultural lending was an attractive prospect for commercial
banks. When the Enfield Committee (UK 1926: 29) on agricultural credit
reported in 1926, it felt able to claim that ‘particularly in the dairy and stockraising districts . . . the accommodation offered by the banks is adequate for
the short-term needs of farmers’. The committee found that in 1926 UK
pounds, 51 million had been lent to farmers, of which just over half (26 million) was for the purchase of agricultural land. In other words, the commercial
banks were funding agricultural land purchases on an extensive scale. Nevertheless, the committee concluded that there was a need for a more uniform
system of agricultural credit with more certainty about the duration of the
loan and less risk of foreclosure. This recommendation led to the establishment of the Agriculture Mortgage Corporation in 1928, a refinancing facility
owned by the clearing banks and the Bank of England. Its direct contacts with
farmers were minimal.
In the postwar period, because British agriculture was already organized
on commercial lines, there was less pressure for state intervention in sup-
port of structural change. The 1947 Agriculture Act was designed to boost
production to meet food security objectives. By increasing farm incomes,
it raised the attractiveness of making loans to farmers, e.g., to tenants to
buy their properties. Marsden et al. (1990: 46) argue that in the UK ‘state
intervention during the postwar period – particularly through guaranteeing
prices and stimulating infrastructural development within agriculture – has
promoted the demand for formal credit links on the part of producers’.
Eventually, the increasing involvement of banks in the agricultural sector
led them to create specialist departments to market their products. Barclays
and Midland were the first two banks to open an agricultural department
around 1970, with other banks following their lead. Hence, in the UK, an
increasingly commercially oriented agricultural sector attracted a growing
level of targeted involvement from the banks. The profit orientation of larger
farmers, the boost to farm incomes given by government subsidies, and the
willingness of the banks to seize a commercial opportunity by treating agriculture as a special case created a spiral leading in the direction of a more
liberal model.
When the pressures for domestic bank deregulation became more intense
in the UK in the 1970s and 1980s, the banking system was already more
open than its counterparts in the other four countries. The one key banking
market that commercial banks had not entered fully, residential mortgages,
was liberalised in 1982. In this environment, the AMC became an increasing
anomaly. As an institution already owned by the banks, however, its structure
did not pose as much of a political obstacle to market liberal change as did
the state-owned agencies in Australia, Canada, and the USA. Its purchase
by Lloyds Bank in the early 1990s was a relatively easy final step toward
fulfilling the conditions for a market liberal model.
Changes in agricultural credit policies have taken place in all five countries,
resulting in arrangements that have taken on, to varying degrees, a more
market liberal character than those in place a quarter century ago. Related
to this admittedly limited policy convergence are parallel changes in policy
communities. Private financial services firms, particularly commercial banks,
have become central players in agricultural finance policy communities. They
have joined with finance ministry officials, and even some agricultural producer representatives, whose views are not so dissimilar from their own. This
coalition of banks, finance ministries, and large agricultural producers has
pushed for more market liberal policies in agricultural finance.
Our proposed avenues for explanations of variations in policy convergence relied less on broad macropolitical institutions than on properties of
policy instruments themselves and their likelihood of reinforcing a path tied
to the developmental model. Specific choices of policy instruments will have
a direct impact on how easy or how difficult it is to change policy at a later
We suggested that the degree to which a state agency created direct interdependent links with farmers, the length of time these links had to take root,
the strata of farmers served by a state agency, and the structural differences
between these state agencies and private firms had an impact on how much
policy convergence occurred. Where linkages were direct and in place for a
long time, and where state agencies were not deposit-taking organizations,
the obstacles to convergence were higher. We found examples of this combination of factors in Australia, Canada, and the USA. In contrast, in Ireland,
links between the ACC and farmers were less deeply rooted and the difference
between the ACC and banks was smaller. In the UK, relationships between
farmers and the AMC were mediated by banks. In these two countries, policy instruments, content, and outcomes converged more fully on the market
liberal model.
We certainly do not deny the importance of macropolitical institutions in
studying policy convergence. We noted the importance of federalism in Australia and Canada. Rather, we conclude that an analysis of these institutions
must be coupled to a micro-level analysis of the institutional characteristics of policy instruments themselves if we are to explain fully variations
among nation-states in policy convergence. And this study shows that even
in a policy sector like agricultural finance where globalisation pressures have
been present, considerable national autonomy remains, resulting in significant
policy divergence.
The authors would like to thank Professor Grace Skogstad, University of
Toronto, for her comments on an earlier draft of this paper. Research supported by the Social Sciences and Humanities Research Council of Canada,
Research Grant 410-92-0260, and by the Economic and Social Research
1. We do not want to say at this stage that the mixed market model is inherently a transitional
one. Some of the evidence we provide below suggests that the model might be a steady
state outcome.
2. After some consolidation with other programmes, it was renamed the Rural Adjustment
Scheme in 1977.
3. Farm Improvement and Market Co-operatives Loans Act.
4. In the reorganization of the US Department of Agriculture in 1994, the FmHA was made
part of a new larger agency, the Consolidated Farm Services Agency.
5. These issues are being explored more fully in a project carried out by Michael Winter at the
Cheltenham and Gloucester College of Higher Education.
Atwood, E. A. (1983). Financing and investment in Irish agriculture, Irish Banking Review,
December 1983.
Australia, Commission of Inquiry into the Australian Financial System (1981), Final Report.
Canberra: AGPS.
Bennett, C. (1991). What is Policy Convergence and What Causes it?, British Journal of
Political Science 21: 215–233.
Bolz, M., Lösch, R. & Meimberg, R. (1993). Agrarkreditsysteme in der Europäischen Union.
Frankfurt aM: Landwirtschaftliche Rentenbank.
Canada, Farm Credit Corporation (1979). The development of farm credit in Canada. Ottawa:
Cerny, P. (1993). The deregulation and re-regulation of financial markets in a more open
world, in Cerny (ed.), Finance and world politics: Markets, regimes and states in the
post-hegemonic era. Aldershot, Hants: Edward Elgar.
Coleman, W. D. (1996). Financial services, globalization, and domestic policy change: A
comparison of North America and the European Union. Basingstoke: Macmillan.
Coleman, W. D., Atkinson, M. & Montpetit, E. (1997). ‘Against the odds’: Retrenchment in
Agriculture in France and the United States, World Politics 49: 453–81.
Coleman, W. D., Skogstad, G. & Atkinson, M. (1996). Paradigm shifts and policy networks:
Cumulative change in agriculture, Journal of Public Policy 16: 273–301.
Coleman, W. D. & Skogstad, G. (1995). Neo-liberalism, policy networks, and policy change,
Australian Journal of Political Science 30: 242–263.
Collins, N. (1995). Agricultural policy networks of the Republic of Ireland and Northern
Ireland, Political Studies 43: 664–682.
Dooney, S. (1988). Irish agriculture: An organisational profile. Dublin: Institute of Public
Feick, J. (1992). Comparing comparative policy studies: A path towards integration?, Journal
of Public Policy 12: 257–285.
Fouilleux, È. (1996). La cogestion à la française à l’épreuve de l’Europe: L’exemple de la
réforme de la Politique Agricole Commune. Grenoble: CERAT-IEP.
Freshwater, D. (1989). The political economy of farm credit reform: The Agricultural Credit
Act of 1987, in C. S. Kramer (ed.), The political economy of US agriculture: Challenges
for the 1990s. Washington, DC: Resources for the Future.
Freshwater, D. & Riemenschneider, C. (1994). Is a revised mandate for the Farm Credit System Needed? Staff paper No. 336, Department of Agricultural Economics, University of
Kentucky, October 1994.
Gifford, J. K., Vivian Wood, J., Reitsma, A. J. & Lougheed, A. L. (1967). Australian Banking,
5th edn. Brisbane: University of Queensland Press.
Grant, W. P. (1995). The limits of Common Agricultural Policy reform and the option of
denationalization, Journal of European Public Policy 2: 1–18.
Grant, W.P. & MacNamara, A. (1995). When policy communities intersect: The case of
agriculture and banking, Political Studies 43: 509–515.
Grant, W. P. & MacNamara, A. (1994). Institutional exceptionalism and the provision of agricultural credit in England and Ireland. Working paper No. 119, Department of Politics and
International Studies, University of Warwick.
Green, G. P. (1987). Finance capital and uneven development. Boulder, CO: Westview Press.
Hamilton, D. E. (1991). From New Day to New Deal: American farm policy from Hoover to
Roosevelt, 1928–1933. Chapel Hill, NC: University of North Carolina Press.
Ives, W. (1987). The Primary Industry Bank of Australia Ltd., 1978–1984. Sydney: PIBA.
Jenkins, S., Flavel, N. & Gargett, D. (1983). Characteristics of rural adjustment scheme
applicants, Quarterly Review of the Rural Economy 5: 371–375.
Lowi, T. J. (1972). Four systems of policy, politics and choice, Public Administration Review
32: 298–310.
Marsden, T., Whatmore, S. & Munton, R. (1990). The role of Banking Capital and Credit
Relations in British Agriculture, in T. Marsden & J. Little (eds.), Political, social and
economic perspectives on the international food system. Aldershot: Avebury.
Maslen, G, (1993). Changes to the rural adjustment scheme. unpublished.
Musgrave, W: (1990). Rural adjustment, in D. B. Williams (ed.), Agriculture in the Australian
economy, 3rd edn. Sydney: Sydney University Press.
OECD, Organisation for Economic Co-operation and Development (1970). Capital and
finance in agriculture, 2 vols. Paris: OECD.
OECD (1991). Ministerial principles for policy reform, Agricultural markets and trade,
monitoring and outlook 1991. Paris: OECD.
Peoples, K. L., Freshwater, D., Hanson, G. D., Prentice, P. T. & Thor, P. (1992). Anatomy of
an American agricultural credit crisis: Farm debt in the 1980s. Lanham, MD: Rowan &
Pierson, P. (1994). Dismantling the welfare state: Reagan, Thatcher and the politics of
retrenchment. New York: Cambridge University Press.
Pierson, P. (1993). When effect becomes cause: Policy feedback and political change, World
Politics 45: 595–628.
Rieger, E. (1996). The common agricultural policy, in H. Wallace & Wallace, W. (eds.), PolicyMaking in the European Union, 3rd edn. Oxford: Oxford University Press.
Scharpf, F. W. (1991). Crisis and choice in European social democracy. Ithaca, NY: Cornell
University Press.
Schickele, R. (1978). Agricultural policy: Farm programs and national welfare. New York:
Skogstad, G. (1990). The farm policy community and public policy in Ontario and Quebec, in
W. D. Coleman & G. Skogstad (eds.), Policy Communities and Public Policy in Canada:
A structural approach. Mississauga: Copp Clark Pitman.
Spillman, K. (1989). Horizons: A history of the Rural and Industries Bank of Western
Australia. Nedlands: University of Western Australia Press.
Standen, B. J. (1982). Credit and agriculture, in D. B. Williams (ed.), Agriculture in the
Australian economy, 2nd edn. Sydney: Sydney University Press.
Stent, W. R. (1981). The changing norms of Australian agricultural policy in the seventies,
Journal of Australian Studies 9: 32–48.
Unger, B. & van Waarden, F. (1995). Introduction: An interdisciplinary approach to convergence, in B. Unger & F. van Waarden (eds.), Convergence or diversity? Internationalization and economic policy response. Aldershot: Avebury.
UK, Ministry of Agriculture and Fisheries (1926). Report on Agricultural Credit. Economic
Series No. 8. London: HMSO.
US Department of Agriculture, Farmers Home Administration (1985). A brief history of
Farmers Home Administration. Washington, DC: USDA.
US General Accounting Office (1992). Farmers Home Administration: Billions of dollars in
farm loans are at risk. Washington, DC: USGAO.
van Waarden, F. (1995). Persistence of national policy styles: A study of their institutional foundations, in B. Unger & F. van Waarden (eds.), Convergence or diversity?
Internationalization and economic policy response. Aldershot: Avebury.
Watson, A. (1989). Rural policies, in A. Patience & B. Head (eds.), From Whitlam to Fraser:
Reform and reaction in Australian Politics. Melbourne: Oxford University Press.
Zysman, J. (1994). Book review of Paulette Kurzer, Business and Banking: Political Change
and Economic Integration in Western Europe [Ithaca, NY: Cornell University Press, 1993],
American Political Science Review 88: 802.
Address for correspondence: Professor William D. Coleman, Department of Political Science,
McMaster University, Hamilton, Ontario, Canada L8S 4M4
Phone: (905) 525-9140, ext. 24741; Fax: (905) 527 3071; E-mail: [email protected]

Policy convergence and policy feedback: Agricultural finance 225 WILLIAM D. COLEMAN